Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method comprising: at a network controller, obtaining data associated with a performance of a plurality of access points of a wireless network, wherein the data comprises first data points associating each particular access point of the plurality of access points with a corresponding performance metric at a particular time; clustering the first data points into a first plurality of groups based on the performance metric to associate each particular access point in the plurality of access points with a corresponding group of the first plurality of groups at the particular time; determining a length of time each particular access point is associated with each group in the first plurality of groups to generate second data points, the second data points indicating a percentage of an overall time interval that each particular access point is associated with each corresponding group of the first plurality of groups; clustering the second data points into a second plurality of groups based on the percentage of the overall time interval each particular access point is associated with each corresponding group in the first plurality of groups, wherein each group in the second plurality of groups associates each particular access point with a relative performance level; selecting a low performance group from the second plurality of groups based on the relative performance level, the low performance group comprising one or more low performance access points; identifying one or more access point configuration changes to increase the performance metric of at least one corresponding low performance access point of the one or more low performance access points; and automatically executing the one or more access point configuration changes on the at least one corresponding low performance access point to improve the performance metric associated with the corresponding low performance access point.
2. The method of claim 1 , wherein the performance metric is a client onboarding failure rate, a radio interference level, or a poor coverage metric.
This invention relates to wireless network optimization, specifically improving network performance by monitoring and adjusting network configurations based on specific performance metrics. The method involves identifying and addressing issues that degrade network quality, such as client onboarding failures, radio interference, or poor coverage. The system continuously evaluates network performance using predefined metrics, including client onboarding failure rates, which measure the percentage of devices that fail to connect to the network successfully. Radio interference levels assess the impact of external signals disrupting network operations, while poor coverage metrics identify areas with weak signal strength or unreliable connectivity. By analyzing these metrics, the system detects performance degradation and automatically adjusts network parameters, such as transmit power, channel selection, or antenna configurations, to mitigate the issues. The method ensures that network performance remains consistent and reliable by dynamically responding to real-time conditions. This approach reduces downtime, enhances user experience, and optimizes resource utilization in wireless networks. The system may also log performance data for further analysis, enabling long-term improvements in network stability and efficiency.
3. The method of claim 1 , further comprising determining that at least one of the first data points is an outlier data point and discarding the outlier data point.
This invention relates to data processing, specifically methods for handling and analyzing datasets to improve accuracy and reliability. The problem addressed is the presence of outlier data points in datasets, which can skew analysis, reduce accuracy, and lead to incorrect conclusions. Outliers are data points that deviate significantly from the rest of the dataset, often due to errors, anomalies, or rare events. The method involves processing a dataset containing multiple data points, where the data points are analyzed to identify and remove outliers. The process begins by evaluating the dataset to detect any data points that fall outside expected ranges or statistical thresholds. Once identified, these outlier data points are discarded from the dataset. This step ensures that subsequent analysis is performed on a cleaned dataset, improving the reliability of results. The method may also include additional steps such as normalizing the data, applying statistical filters, or using machine learning techniques to refine the outlier detection process. By removing outliers, the method enhances the accuracy of data-driven decisions, making it particularly useful in fields like finance, healthcare, and engineering, where precise data analysis is critical. The invention provides a systematic approach to data cleaning, ensuring that only relevant and reliable data points are used for further processing.
4. The method of claim 1 , wherein clustering the first data points comprises k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering.
This invention relates to data clustering techniques used in machine learning and data analysis. The problem addressed is the need for efficient and accurate methods to group similar data points in a dataset, which is essential for tasks like pattern recognition, anomaly detection, and data compression. The invention describes a method for clustering data points by applying one or more clustering algorithms to a dataset. The clustering process involves selecting a subset of data points from the dataset and applying a clustering algorithm to group these points into clusters based on their similarity. The clustering algorithms used include k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering. Each algorithm has distinct advantages depending on the data structure and the specific requirements of the analysis. For example, k-means clustering is efficient for spherical clusters, while density-based methods like DBSCAN are better for identifying clusters of varying shapes and handling noise. The method allows for flexibility in choosing the most appropriate clustering technique based on the characteristics of the data and the desired outcome. This approach enhances the accuracy and adaptability of data clustering in various applications, such as customer segmentation, image processing, and bioinformatics.
5. The method of claim 1 , wherein clustering the second data points comprises k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering.
This invention relates to data clustering techniques used in data analysis and machine learning. The problem addressed is the need for efficient and accurate clustering of data points to identify patterns, groupings, or structures within large datasets. Traditional clustering methods may struggle with scalability, noise, or complex data distributions, leading to suboptimal results. The invention describes a method for clustering data points using various clustering algorithms. The method involves applying one or more clustering techniques to a set of data points to group them into clusters based on similarity. The supported clustering algorithms include k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, and agglomerative clustering. Each algorithm has distinct advantages depending on the data characteristics, such as the number of clusters, data density, or noise levels. For example, k-means is efficient for spherical clusters, while density-based methods handle irregularly shaped clusters. The method may be applied in fields like image segmentation, customer segmentation, or anomaly detection, where identifying meaningful groupings in data is critical. The invention improves clustering accuracy and adaptability by allowing the selection of the most suitable algorithm for the given dataset.
6. The method of claim 1 , wherein the one or more access point configuration changes includes changing a transmission power of the corresponding low performance access point or changing a channel assignment of the corresponding low performance access point.
This invention relates to wireless network optimization, specifically improving performance in environments with multiple access points (APs) by dynamically adjusting configurations of low-performance APs. The problem addressed is inefficient network performance due to suboptimal AP settings, such as excessive interference, poor signal coverage, or channel congestion, which degrade user experience and network efficiency. The method involves monitoring network performance metrics, such as signal strength, data throughput, or client connectivity, to identify low-performance APs. Once identified, the method adjusts the configuration of these APs to mitigate performance issues. Specifically, the adjustments include modifying the transmission power of the low-performance AP to reduce interference or improve coverage, or changing the channel assignment to avoid congestion or conflicts with neighboring APs. These changes are made automatically based on real-time or historical performance data, ensuring continuous optimization without manual intervention. By dynamically adjusting AP configurations, the invention enhances network reliability, reduces interference, and improves overall user experience in dense or complex wireless environments. The solution is particularly useful in scenarios where static configurations fail to adapt to changing network conditions or user demands.
7. The method of claim 1 , further comprising: obtaining updated data associated with the performance of the plurality of access points; and recursively updating the first plurality of groups, the second plurality of groups, the low performance group, and the one or more access point configuration changes.
This invention relates to wireless network optimization, specifically improving the performance of access points (APs) in a network by dynamically grouping and configuring them based on performance data. The problem addressed is the static or inefficient grouping of APs, which can lead to suboptimal performance, coverage gaps, or interference. The solution involves dynamically organizing APs into multiple groups based on their performance metrics, such as signal strength, throughput, or latency. A first set of groups is formed to optimize coverage, while a second set is formed to optimize capacity. APs that consistently underperform are identified and placed in a low-performance group for targeted adjustments. The system then determines and applies configuration changes to the APs, such as adjusting transmit power, channel selection, or load balancing settings, to improve overall network efficiency. The process is recursive, continuously updating the groupings and configurations as new performance data is collected, ensuring the network adapts to changing conditions. This approach enhances network reliability, reduces manual intervention, and improves user experience by dynamically optimizing AP performance.
8. An apparatus comprising: a network interface configured to communicate with a plurality of access points in a wireless network; and a processor coupled to the network interface and configured to: obtain via the network interface, data associated with a performance of the plurality of access points of the wireless network, wherein the data comprises first data points associating each particular access point of the plurality of access points with a corresponding performance metric at a particular time; cluster the first data points into a first plurality of groups based on the performance metric to associate each particular access point in the plurality of access points with a corresponding group of the first plurality of groups at the particular time; determine a length of time each particular access point is associated with each group in the first plurality of groups to generate second data points, the second data points indicating a percentage of an overall time interval that each particular access point is associated with each corresponding group of the first plurality of groups; cluster the second data points into a second plurality of groups based on the percentage of the overall time interval each particular access point is associated with each corresponding group in the first plurality of groups, wherein each group in the second plurality of groups associates each particular access point with a relative performance level; select a low performance group from the second plurality of groups based on the relative performance level, the low performance group comprising one or more low performance access points; identify one or more access point configuration changes to increase the performance metric of at least one corresponding low performance access point of the one or more low performance access points; and automatically execute the one or more access point configuration changes on the at least one corresponding low performance access point to improve the performance metric associated with the corresponding low performance access point.
This invention relates to wireless network optimization by analyzing and improving access point performance. The system monitors multiple access points in a wireless network, collecting performance data over time. Each access point is evaluated based on performance metrics, and the data points are grouped into clusters representing different performance levels at specific times. The system then calculates the duration each access point remains in each performance cluster, generating a percentage of time spent in each cluster. These percentages are used to assign access points to new clusters based on their relative performance over time. The system identifies low-performing access points by selecting the cluster with the worst performance levels. To address performance issues, the system determines configuration changes, such as adjusting transmission power or channel settings, and automatically applies these changes to the low-performing access points. This approach dynamically optimizes network performance by continuously analyzing and adjusting access point configurations based on historical and real-time data.
9. The apparatus of claim 8 , wherein the performance metric is a client onboarding failure rate, a radio interference level, or a poor coverage metric.
This invention relates to wireless communication systems, specifically addressing the challenge of optimizing network performance by monitoring and adjusting key metrics. The apparatus includes a processor and a memory storing instructions that, when executed, cause the processor to collect performance data from a wireless network. The collected data is analyzed to determine whether a predefined performance threshold has been exceeded. If the threshold is exceeded, the apparatus triggers an automated adjustment of network parameters to improve performance. The performance metrics being monitored include client onboarding failure rates, radio interference levels, and poor coverage metrics. These metrics help identify issues such as failed device connections, excessive interference, or inadequate signal strength, which can degrade network reliability and user experience. By continuously monitoring these metrics and automatically adjusting network settings, the apparatus ensures optimal network operation without manual intervention. The system is designed to enhance network efficiency, reduce downtime, and improve overall service quality in wireless communication environments.
10. The apparatus of claim 8 , wherein the processor is further configured to determine that at least one of the first data points is an outlier data point and discard the outlier data point.
This invention relates to data processing systems that analyze datasets to identify and remove outlier data points. The problem addressed is the presence of anomalous or erroneous data points in datasets, which can skew analysis results and reduce the accuracy of data-driven decisions. The apparatus includes a processor configured to process a dataset containing multiple data points, where the processor identifies and discards outlier data points to improve data quality. The processor evaluates the dataset to detect outliers, which are data points that deviate significantly from the majority of the data. Once identified, these outliers are removed from the dataset to ensure that subsequent analysis is based on more reliable and representative data. The apparatus may also include a memory for storing the dataset and a display for visualizing the processed data. The outlier detection and removal process enhances the accuracy of statistical analyses, machine learning models, and other data-driven applications by eliminating erroneous or anomalous values that could distort results. This approach is particularly useful in fields such as finance, healthcare, and engineering, where data integrity is critical.
11. The apparatus of claim 8 , wherein the processor is configured to cluster the first data points by k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering.
This invention relates to data clustering techniques used in data analysis and machine learning. The problem addressed is the need for flexible and efficient clustering methods to group data points into meaningful clusters based on their characteristics. Traditional clustering algorithms may lack adaptability to different data distributions or may not effectively handle noise and outliers. The apparatus includes a processor configured to perform clustering on a set of first data points. The processor applies one or more clustering algorithms to group the data points into clusters. Specifically, the processor can use k-means clustering, which partitions data into k predefined clusters by minimizing within-cluster variance. Alternatively, mean-shift clustering can be employed, which identifies clusters by finding dense regions in the data space. Density-based spatial clustering is another option, which groups together points that are close together in high-density regions while marking outliers as noise. Expectation maximization clustering can also be used, which models data as a mixture of Gaussian distributions and iteratively refines cluster assignments. Finally, agglomerative clustering is supported, which builds clusters by iteratively merging the closest pairs of clusters. The processor selects the appropriate clustering method based on the data characteristics and the desired clustering outcome, ensuring robust and adaptable data analysis.
12. The apparatus of claim 8 , wherein the processor is configured to cluster the second data points by k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering.
This invention relates to data processing systems that analyze and cluster data points to identify patterns or groupings within a dataset. The problem addressed is the need for efficient and accurate clustering of data points to reveal meaningful structures in large datasets, which is essential for applications such as machine learning, data mining, and pattern recognition. The apparatus includes a processor configured to perform clustering operations on a set of second data points, which are derived from an initial set of first data points. The processor applies one or more clustering algorithms to the second data points to group them into clusters based on similarity or proximity. The clustering algorithms may include k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering. Each of these algorithms has distinct advantages depending on the dataset characteristics, such as the number of clusters, data distribution, or noise levels. K-means clustering partitions the data into k predefined clusters by iteratively minimizing the distance between data points and cluster centroids. Mean-shift clustering identifies clusters by shifting data points toward regions of higher density. Density-based spatial clustering groups data points based on density connectivity, allowing for the detection of arbitrarily shaped clusters. Expectation maximization clustering models the data as a mixture of probability distributions and iteratively refines the model parameters. Agglomerative clustering builds a hierarchy of clusters by iteratively merging the closest pairs of clusters. The apparatus enables flexible and adaptive clustering of data points, improving the accuracy and efficiency of data analysis
13. The apparatus of claim 8 , wherein the one or more access point configuration changes includes changing a transmission power of the corresponding low performance access point or changing a channel assignment of the corresponding low performance access point.
This invention relates to wireless network optimization, specifically improving network performance by dynamically adjusting configurations of low-performance access points (APs). The problem addressed is inefficient use of network resources due to suboptimal AP settings, leading to poor connectivity, congestion, or dead zones. The solution involves monitoring network conditions and automatically modifying AP configurations to enhance overall performance. The apparatus includes a network analyzer that identifies low-performance APs based on metrics such as signal strength, throughput, or client device connectivity. Once identified, the system adjusts the AP's transmission power or channel assignment to mitigate performance issues. Reducing transmission power can minimize interference with neighboring APs, while reassigning channels can avoid crowded frequencies. These changes are applied selectively to low-performance APs, ensuring high-performance APs remain unaffected. The system may also revert changes if performance does not improve, ensuring stability. This dynamic adjustment optimizes network efficiency without manual intervention, improving user experience and resource utilization.
14. The apparatus of claim 8 , wherein the processor is further configured to: obtain updated data associated with the performance of the plurality of access points; and recursively update the first plurality of groups, the second plurality of groups, the low performance group, and the one or more access point configuration changes.
This invention relates to wireless network optimization, specifically improving the performance of access points (APs) in a network by dynamically grouping and adjusting their configurations. The problem addressed is inefficient network performance due to suboptimal AP configurations, leading to poor connectivity, coverage gaps, or congestion. The solution involves a system that analyzes AP performance data, categorizes APs into groups based on their performance metrics, and applies configuration changes to optimize network efficiency. The system includes a processor that obtains performance data from multiple APs, such as signal strength, throughput, or latency. The processor then groups the APs into a first set of high-performance groups and a second set of low-performance groups. APs in the low-performance group are further analyzed to identify specific configuration changes needed to improve their performance. These changes may include adjusting transmission power, channel selection, or load balancing. The processor recursively updates the groupings and configuration changes as new performance data is obtained, ensuring continuous optimization. This dynamic approach allows the network to adapt to changing conditions, such as user demand or environmental interference, without manual intervention. The result is a more reliable and efficient wireless network with minimized performance degradation.
15. One or more non-transitory computer readable storage media encoded with software comprising computer executable instructions and, when the software is executed by a processor on a network controller, operable to cause the processor to: obtain data associated with a performance of a plurality of access points of a wireless network, wherein the data comprises first data points associating each particular access point of the plurality of access points with a corresponding performance metric at a particular time; cluster the first data points into a first plurality of groups based on the performance metric to associate each particular access point in the plurality of access points with a corresponding group of the first plurality of groups at the particular time; determine a length of time each particular access point is associated with each group in the first plurality of groups to generate second data points, the second data points indicating a percentage of an overall time interval that each particular access point is associated with each corresponding group of the first plurality of groups; cluster the second data points into a second plurality of groups based on the percentage of the overall time interval each particular access point is associated with each corresponding group in the first plurality of groups, wherein each group in the second plurality of groups associates each particular access point with a relative performance level; select a low performance group from the second plurality of groups based on the relative performance level, the low performance group comprising one or more low performance access points; identify one or more access point configuration changes to increase the performance metric of at least one corresponding low performance access point of the one or more low performance access points; and automatically execute the one or more access point configuration changes on the at least one corresponding low performance access point to improve the performance metric associated with the corresponding low performance access point.
The invention relates to optimizing wireless network performance by analyzing and adjusting access point configurations. The problem addressed is the need to dynamically improve the performance of access points in a wireless network based on their operational data. The system obtains performance data for multiple access points, where each data point links an access point to a performance metric at a specific time. These data points are clustered into groups based on performance metrics, associating each access point with a group at that time. The system then calculates the duration each access point remains in each group, generating new data points that represent the percentage of time an access point spends in each group. These new data points are clustered again to assign each access point a relative performance level. A low-performance group is identified, and configuration changes are proposed and automatically applied to the low-performing access points to enhance their performance metrics. The solution automates the detection and correction of underperforming access points, improving overall network efficiency.
16. The non-transitory computer readable storage media of claim 15 , wherein the performance metric is a client onboarding failure rate, a radio interference level, or a poor coverage metric.
This invention relates to wireless network optimization, specifically improving network performance by analyzing and mitigating issues such as client onboarding failures, radio interference, and poor coverage. The system monitors network performance using predefined metrics, including client onboarding failure rates, radio interference levels, and coverage quality indicators. When performance degradation is detected, the system automatically adjusts network parameters to mitigate the issue. For example, it may modify transmission power, channel selection, or handover thresholds to reduce interference or improve coverage. The system also logs performance data for historical analysis and trend detection, allowing for proactive adjustments. The solution is designed for wireless networks, including cellular and Wi-Fi systems, to enhance reliability and user experience by dynamically responding to real-time performance metrics. The invention focuses on automated, data-driven optimization to minimize manual intervention while maintaining network efficiency.
17. The non-transitory computer readable storage media of claim 15 , further comprising instructions operable to cause the processor to determine that at least one of the first data points is an outlier data point and discard the outlier data point.
The invention relates to data processing systems that analyze datasets to identify and remove outlier data points. In many applications, datasets contain anomalies or outliers that can skew analysis results, reduce accuracy, or lead to incorrect conclusions. The invention addresses this problem by providing a method to automatically detect and discard outlier data points from a dataset before further processing. The system processes a dataset containing multiple data points, each associated with a value. It calculates statistical measures, such as mean, median, or standard deviation, to assess the distribution of the data. Using these measures, the system identifies data points that deviate significantly from the expected range or pattern, classifying them as outliers. Once identified, these outliers are removed from the dataset to improve the reliability of subsequent analysis. The invention may be implemented in software, hardware, or a combination of both, and can be applied in various fields such as finance, healthcare, manufacturing, and scientific research, where accurate data analysis is critical. By automatically detecting and discarding outliers, the system enhances data quality and ensures more accurate and trustworthy results.
18. The non-transitory computer readable storage media of claim 15 , further comprising instructions operable to cause the processor to cluster the first data points by k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering.
This invention relates to data processing systems that analyze and cluster data points using machine learning techniques. The problem addressed is the need for efficient and accurate clustering of data points in large datasets to identify patterns, groupings, or anomalies. The invention provides a computer-implemented method for processing data, where a processor executes instructions stored on a non-transitory computer-readable storage medium. The method involves receiving a dataset containing multiple data points, each associated with one or more features. The processor then applies a clustering algorithm to group the data points into clusters based on their feature similarities. The clustering algorithms include k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering. These algorithms are selected based on their ability to handle different data distributions and structures, ensuring robust and adaptable clustering performance. The method may also include preprocessing steps to normalize or transform the data before clustering, as well as post-processing steps to refine or validate the resulting clusters. The invention aims to improve the accuracy and efficiency of data analysis by leveraging multiple clustering techniques, allowing users to choose the most suitable method for their specific dataset.
19. The non-transitory computer readable storage media of claim 15 , further comprising instructions operable to cause the processor to cluster the second data points by k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering.
This invention relates to data processing systems that analyze and cluster data points using machine learning techniques. The problem addressed is the need for efficient and accurate clustering of data points in large datasets to identify patterns, groupings, or anomalies. The system processes a dataset containing first data points and second data points, where the second data points are derived from the first data points through a transformation process. The transformation may involve dimensionality reduction, feature extraction, or other preprocessing steps to prepare the data for clustering. The system then applies clustering algorithms to the second data points to group them into meaningful clusters. The clustering algorithms include k-means clustering, mean-shift clustering, density-based spatial clustering, expectation maximization clustering, or agglomerative clustering. These algorithms are selected based on their ability to handle different types of data distributions and noise levels. The clustering results are used to derive insights, such as identifying trends, detecting outliers, or segmenting the data for further analysis. The system is designed to improve the accuracy and efficiency of data clustering in applications such as customer segmentation, anomaly detection, and pattern recognition.
20. The non-transitory computer readable storage media of claim 15 , wherein the one or more access point configuration changes includes changing a transmission power of the corresponding low performance access point or changing a channel assignment of the corresponding low performance access point.
Wireless network optimization systems often struggle with balancing performance across multiple access points (APs), particularly when some APs exhibit lower performance due to interference, congestion, or suboptimal configurations. This invention addresses the problem by dynamically adjusting configurations of low-performance APs to improve overall network efficiency. The system identifies APs with performance issues, such as high latency or low throughput, and applies targeted changes to their settings. These changes may include modifying transmission power levels to reduce interference or reassigning channel assignments to avoid overlapping frequencies with neighboring APs. By selectively adjusting these parameters, the system enhances signal quality and reduces congestion, leading to better performance for connected devices. The solution leverages real-time monitoring and automated adjustments to maintain optimal network conditions without manual intervention. This approach is particularly useful in dense wireless environments where static configurations often lead to inefficiencies. The invention ensures that low-performance APs are dynamically reconfigured to mitigate their impact on the broader network, improving reliability and user experience.
Unknown
September 17, 2019
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.